Pengelompokan Pelanggan Supermarket Berdasarkan Riwayat Transaksi Dengan Metode K-Means Clustering

Abstract

This research aims to apply the K-Means Clustering method to group supermarket customers based on their transaction history. Amid the increasingly competitive retail business landscape and rapid changes in consumer behavior, a deep understanding of customer segmentation is crucial for enhancing profitability and maintaining customer loyalty. As business entities, supermarkets generate a large volume of transaction data daily. This data holds valuable information that can be analyzed to uncover hidden patterns in purchasing behavior, product preferences, as well as customer visit frequency and transaction value. The K-Means Clustering method was chosen due to its algorithmic simplicity and efficiency in handling large-scale data. This method partitions customer data into several clusters based on similarities in certain characteristics, such as total purchases, visit frequency, and time of last transaction. The clustering results are expected to identify various customer segments, such as loyal customers, at-risk customers, and potential customers whose loyalty can be enhanced through targeted approaches. The insights gained from this grouping process will be valuable for management in designing more personalized and targeted marketing strategies, developing appropriate loyalty programs, and optimizing stock management and product promotions. Therefore, this research provides a significant contribution by offering a scientific foundation for data-driven strategic decision-making, particularly in customer management within the supermarket retail sector.